Published February 1, 1988
Changes in the Expression of Collagen Genes Show Two Stages in
Chondrocyte Differentiation In Vitro
Patrizio Castagnola, Beatrice Dozin, Giorgia Moro, and Ranieri Cancedda
Laboratorio di Differenziamento Cellulare, Istituto Nazionale per la Ricerca sul Cancro, Genoa, Italy
Abstract. This report deals with the quantitation of
course of the culture, reached its maximal value after
3-4 wk, and decreased only at a later stage of cell
differentiation.
As determined by in vitro run-off transcription assays, all these changes in collagen mRNA levels could
be attributed to parallel modifications in the relative
rate of transcription of the corresponding collagen
genes.
We suggest that chicken chondrocyte differentiation
proceeds through at least two different steps: (a) first,
transition from a stage characterized by a high level of
type I collagen mRNA to a stage characterized by predominance of types II and IX collagen mRNAs; (b)
later, transition to a stage characterized by the highest
level of type X collagen mRNA.
RGANOGENESlSof long bones involves cellular differentiation and continuous synthesis and remodeling of
the extracellular matrix. During the first stages of
chick development, mesenchymal cells in the limb buds
differentiate, possibly through distinct regulatory steps (22),
and condense to form a blastema from which cartilaginous
bones develop. This process is characterized by profound
changes in the cell morphology accompanied by the synthesis of novel extracellular macromolecules, among which are
cartilage-specific proteoglycans and collagens (7). Changes
in the levels of types I and II collagens and of their mRNAs
during chondrogenesis in vivo and in vitro were detected
with the use of specific antibodies and cDNA probes (27, 10,
11). Cartilaginous bones grow because of cell proliferation
and deposition of new extracellular matrix. Proliferating
chondrocytes progressively mature into hypertrophic nondividing cells. Hypertrophy of chondrocytes, first detectable
in the diaphysis and after in the epiphyseal regions, is accompanied by the onset of type X collagen synthesis (2, 20).
Hypertrophic cartilage undergoes calcification, is invaded by
blood vessels and osteogenic cells, and is replaced by bone
tissue (18). To investigate at the molecular level the differentiation of hypertrophic cartilage chondrocytes we have used an
in vitro system starting with tibial chondrocytes of early stage
chick embryos (3). In this system, chondrocytes obtained by
enzymatic dissociation of stage 28-30 (9) embryo tibiae,
passaged on tissue culture dishes, assume a fibroblast-like
morphology and switch from type II to type I collagen synthesis. When these adherent cells are transferred to agarosecoated dishes (suspension culture) they revert to the chondrocytic phenotype, and, after a transient stage in which cells
form aggregates, they mature to single hypertrophic chondrocytes that produce collagen type X (3).
We have also shown that the addition of ascorbic acid to
the suspension culture results in the organization of the extracellular matrix and in the maturation of the cell aggregates
into hypertrophic cartilage (23). In this paper we report the
variations in the steady-state levels of types I, II, IX, and X
collagen mRNAs and the transcriptional activity of the corresponding genes during the in vitro differentiation of chick
embryo tibial chondrocytes starting from adherent cells that
produce type I collagen. As in in vivo studies on mRNA levels during limb bud development (10, 11), we have found a
rapid decrease of the type I collagen mRNA and a concomitant and rapid increase of the type II collagen mRNA in the
first 2 wk of culture. At the same time the mRNA for the type
IX collagen increased. The increase of the type X collagen
mRNA was more progressive and continuous, and at later
times of culture was simultaneous to the decrease of types
II and IX collagen mRNAs. All these changes in collagen
mRNA levels could be correlated to parallel changes in the
transcription rate of the specific genes. These findings lead
us to suggest that, during the process of hypertrophic chondrocyte differentiation, a shift in production from type I to
9 The Rockefeller University Press, 0021-9525/88/02/461/7 $2.00
The Journal of Cell Biology, Volume 106, February 1988 461-467
461
Downloaded from on June 16, 2017
both mRNA and transcription activity of type I collagen gene and of three cartilage-specific collagens
(types II, IX, and X) during in vitro differentiation of
chick chondrocytes. Differentiation was obtained by
transferal to suspension culture of dedifferentiated cells
passaged for 3 wk as adherent cells. The type I collagen mRNA, highly represented in the dedifferentiated
cells, rapidly decreased during chondrocyte differentiation. On the contrary, types II and IX collagen mRNAs
sharply increased within the first week of suspension
culture, peaked in the second week, and thereafter began to decrease. This decrease was particularly significant for type IX collagen mRNA. The level of type X
collagen mRNA progressively increased during the
Published February 1, 1988
types II and IX collagens occurs when chondrogenesis begins, and a shift from types II and IX to type X collagens
occurs when chondrocytes become hypertrophic.
Materials and Methods
Cell Culture
Adherent dedifferentiated and differentiating chondrocytes were obtained as
described by Castagnola et al. (3). Briefly, dedifferentiated chondrocytes
were obtained by plating freshly dissociated choodrocytes from tibiae of
stage 28-30 chick embryos (9) in plastic tissue culture dishes.
Differentiating chondrocytes were obtained by transferring the dedifferentiated cells cultured for 3 wk till adherent to the plastic dish to 1% agarose-coated dishes (suspension culture). For the experiments presented in
Fig. 4, chondrocytes were dissociated from tibiae of stage 30 chick embryo
and cultured directly in suspension condition.
Culture medium was Coon's modified F12 medium (1) lacking ascorbic
acid and supplemented with 10% FCS.
Origin and Labeling of eDNA Probes
RNA Extraction and Hybridization Analysis
RNA was extracted by guanidinium isothiocyanate/CsCl density gradient
centrifugation, denatured, and subjected to electrophoresis through 1%
agarose gels containing formaldehyde or glyoxal according to the procedure
described by Maniatis et al. (13). Northern blots were performed by capillarity on a nylon membrane (Hybond; Amersham Corp., Arlington
Heights, IL). RNA slot blots were performed with a manifold apparatus
(Schleicher & Schuell, Inc.) after denaturation at 65~ for 15 rain in 4 x
SSC (lx SSC solution is 0.15 M sodium chloride, 0.015 M sodium citrate)
containing 25.9 % formaldehyde, and chilling on ice for 15 min. (24). Denatured RNA was spotted at concentrations of •20, 5, and 1.25 ~tg on a Hybond nylon membrane.
Binding of RNA to membrane and hybridization were carried out according to the directions of the membrane manufacturer with the following
modifications: hybridization and washing temperature was 54~ filters
were washed four times in l• SSPE (20x SSPE solution is 3.6 M NaCI,
0.2 M Na phosphate buffer, pH 7.7, 0.02 M Na2 EDTA) containing 0.2%
SDS for 10 min and then twice in 2 x SSPE, 0.2% SDS for 30 rain to avoid
cross-hybridization (6). In addition, when the probe for al (II) was used,
the formamide concentration was raised from 50 to 60% vol/vol to increase
hybridization specificity. When the filters were reused with different probes
they were washed, hybridized for 16 h at 74~ in 5 mM Tris-HCl, pH 8.0,
2 mM Na2 EDTA, 1• Denhardt's solution (100x solution is 2% wt/vol
BSA, 2 % wt/vol fieol170, 2 % wt/vol potyvinylpyrrolidone), and autoradiographed to ensure that no probe remained on the filters. Filters were autoradiographed wet for different times (3-72 h) at - 8 0 ~ using intensifying
screens. Different exposures were scanned at 590 nm with a spectrophotometer (DU8; Beckman Instruments Inc., Palo Alto, CA).
Results
Collagen mRNA Levels during In Vitro Differentiation
of Chick Embryo Chondrocytes
CelLs were collected, washed once in PBS (pH Z2), and resuspended at a
concentration of 5 • l& cells/ml in buffer A (10 mM Tris-HC1, pH 7.6,
40% glycerol, 10 mM MgCI2, 10 mM NaCI) containing 0.1% Triton X-100,
and precooled at -20~ Cells were then homogenized with 35 strokes of
a tightly fitted Dounce homogenizer. Nuclei were pelleted by centrifugation
To avoid uncontrolled loss during fractionation, total RNA
was used for analysis. Total RNA was extracted from
dedifferentiated chondrocytes grown for 3 wk as adherent
cells on tissue culture plastic dishes and at different time intervals after their transfer to suspension culture on agarosecoated dishes. Aliquots of the extracts were analyzed both by
Northern and slot blots with probes for the al chains of types
I, II, and X collagens and the a2 chain of type IX collagen
(Figs. 1 and 2). To minimize experimental errors, each hybridized filter was exposed for different lengths of time and
only exposures in the proportional range of the autoradiographic film darkening were considered for densitometry
scanning analysis. The amount of RNA was normalized with
respect to the amount of ribosomal RNA as detected by hybridization with the specific probe pXCR7.
The results indicate that immediately after the transfer of
the cells to suspension culture, the level of type I collagen
mRNA rapidly decreased to, by the fourth week, a value
equivalent to 10% of that determined for the dedifferentiated
cells. On the contrary, within the first week, the amount of
types II and IX collagen mRNAs dramatically increased and
reached ~80 and 90%, respectively, of the maximal levels
observed at the second week. At later times of culture the
amount of types II and IX collagen mRNAs decreased. A
different pattern was observed for type X collagen mRNA.
This mRNA accumulated more progressively throughout
culture, and at the end of the first week its concentration was
only 40% of the maximum observed at the fourth week.
The Journal of Cell Biology, Volume 106, 1988
462
In Vitro Run-off Transcription Assay
Downloaded from on June 16, 2017
The eDNA clone pCIII8 (al X) was obtained in our laboratory (4); the
eDNA clones pCOL3 (al I), pCS1 (al II), and pYN 1731 (a2 IX) were
kindly provided by E. Avvedimento (28), M. Sobel (National Institutes of
Health, Bethesda, MD; see reference 29), and Y. Ninomiya (Harvard Medical SchooL, Boston, MA) (17), respectively; the genomic clone pXCR7
(Hind III-Hind III fragment of Xenopus laevis rDNA) was a gift from
E Amaldi (Universita Tor Vergata, Rome, Italy).
Probes were labeled by the standard nick translation method as described
by Maniatis et al. (12). The specific activity of the probes was always >10s
cpm/l~g of DNA. Whole plasmids were used as probes only when filters
were hybridized for the first time. To avoid possible cross-reaction when the
filters were washed and reused for hybridization with different probes, only
the eDNA inserts were used after excision from the vectors by restriction
enzymes and purification on an agarose gel/NA 45 membrane (Schleicher
& Schuell, Inc., Keene, NH).
at 4,000 g for 10 min at -15~ washed twice in the same buffer without
Triton, purified by centrifngation at 150,000 g for 65 rain at 40C through
a 1.8-M sucrose cushion in 10 mM Tris-HCI, pH 7.6, and 10 mM MgCI2,
and stored in buffer A at -80*C at a concentration of 1-3 x 10S/ml. Transcription assays were performed according to the method of Clayton and
Darnell (5) with the following modifications: nuclei were adjusted to 15-25
absorbance units at 260 nm (determined in 1% SDS), and the reaction
medium contained a 1-mM concentration each of ATE GTP, and CTP, 0.5
U/ml of RNase inhibitor (Boehringer Mannheim), and 125 ~tCi [a-32p]UTP
(3,000 Ci/mmol; Amersham, Corp.). When assaying RNA polymerase 1I
activity, nuclei were preineubated on ice for 10 min in the same buffer supplemented with 1 ~ / m l alpha-amanitin. Transcription incubation was cartied out for 30 rain at 30*C, the time interval that, in preliminary experiments, allowed an optimal incorporation of [32p]UTP to occur at that
temperature (data not shown). The reaction was terminated by digestion
with 100 t~g/ml concentrations each of DNase I and proteinase K in the presence of 1 mM CaCI2 (25). Yeast tRNA was added to a final concentration
of 25 gg/ml and RNA was extracted with phenol/chloroform/isoamylalcohol (25:24:1) and precipitated with 0.5 vol of 7.5 M ammonium acetate and
2.5 vol of ethanol. RNA hybridization was carried out in DNA excess conditions (2 ~tg per assay) as described by McKnight and Palmiter (14) with the
collagen eDNA inserts excised from the plasmids described above and
immobilized on nitrocellulose filters. The length of these eDNA restriction
fragments were: Pst I-Pst I, 530 bp for pCIII8; Hind Ill-Hind III, 800 bp
for pCOL3; Pst I-Pst I, 780 bp for pCSI; and Hind IlI-Rsa I, 1,600 bp for
pYN1731. In each assay, nonspecific hybridization was estimated in parallel
with a filter devoid of eDNA. After a 3-d hybridization, the filters were
washed and digested with RNases A and T~ (1 klg/ml and 10 U/ml, respectively) (5). The [32P]RNAs were eluted (14) and the radioactivity was
quantitated in Pico-Fluor 40 (Packard Instrument Co., Downer's Grove,
IL). The relative rate of a gene transcription for each type of collagen is
expressed as parts per million (ppm), which is the ratio between counts per
minute specifically hybridized to the corresponding eDNA and counts per
minute • 10"~ in total transcription.
Published February 1, 1988
Downloaded from on June 16, 2017
Figure 2. Slot blots of total RNAs extracted from chondrocytes
grown for 3 wk on plastic dishes (lanes 0) and from the same cells
transferred and cultured on agarose-coated dishes for 1, 2, and 4 wk
(lanes 1, 2, and 4). Control RNA (lanes C) was from human lymphoid T cell line. Probes used are indicated on top of each panel.
Amount of loaded RNA was ~20, 5, and 1.25 tig. The filters were
hybridized first with the collagen-specific probe and then with the
rRNA probe. For types I and IX collagen probes the same filter was
used.
Figure1. Northern blots of total RNAs extracted from chondrocytes
grown for 3 wk on plastic dishes (lanes 0) and from the same cells
transferred and cultured on agarose-coated dishes for 1, 2, 3, or
4 wk (lanes 1-4). Control RNA (lanes C) was from chicken liver.
On each slot "~30 ~tg of RNA were loaded. Probes used are indicated at top of each panel. Arrows refer to the position of ribosomal
RNAs. At the bottom of each panel the 18S rRNA region of the
same filter after rehybridization with the pXCR7 probe for rRNA
is shown.
Castagnola et al. CollagenExpressionin DifferentiatingChondrocytes
It must be noted that low but detectable levels of cartilagespecific mRNAs were observed in adherent cells.
Transcriptional Activity of Collagen Genes
in Differentiating Chondrocytes
To investigate whether modifications in the rate of transcription of the collagen genes could account for the changes in
m R N A levels described above, we performed, in an independent experiment, in vitro run-off transcription assays on
nuclei isolated from cells cultured either adherent to the
dishes or grown in suspension for 1-4 wk. Control assays
463
Published February 1, 1988
Table I. Relative Rate of Transcription of the Genes
Encoding Types I, H, IX, and X Collagens during
Chondrocyte Differentiation In Vitro
Collagens
Culture
conditions
Dedifferentiated
(adherent)
Dedifferentiated +
0t-amanitin
1-wk Suspension
2-wk Suspension
2-wk Suspension +
ct-amanitin
3-wk Suspension
4-wk Suspension
Total
transcription
Type I Type II Type IX Type X
cpm x 1 0 -6
ppm
ppm
ppm
24.4
207
2
5
11
11.7
18.7
20.9
45
127
98
0.95
35
36
1.6
74
61
2.3
218
419
12.7
24.0
22.7
18
57
19
12
30
15
14
66
58
88
643
438
ppm
Dedifferentiated chondrocytes cultured as adherent cells for 3 wk were transferred to suspension culture for 1-4 wk. The rate of transcription of the genes
encoding collagen types I, II, IX, and X was measured by the level of incorporation of [32P]UTP into nascent RNA transcripts in isolated nuclei and is expressed as parts per million (ppm) as described in Materials and Methods.
Nonspecific hybridization averaged 100, 11, 19, and 54 cpm for collagen types
I, II, IX, and X, respectively.
GENE
L
~
] TRANIICIIIlPTION
with alpha-amanitin were also included. Transcription data,
expressed as parts per million, are presented in Table I. The
transfer of dedifferentiated adherent cells to a suspension culture allowing cellular differentiation resulted in a marked
and continuous decrease in the transcriptional activity of the
gene encoding collagen type I. In contrast, in the same
differentiating cells, the transcription rate of the genes
specific for collagen types II, IX, and X was significantly enhanced, the highest increase being that for the gene encoding
collagen type X (,x,60-fold stimulation with respect to
dedifferentiated cells). These three genes showed an overall
similar pattern of activation, reaching maximal transcription
rate between the first and second weeks (collagen types II
and IX) or within the third week (collagen type X), and
decreasing thereafter to a lower level at the fourth week of
suspension culture. Finally, all the changes in gene transcription rate described above were specific since (a) the change
of culture conditions (adherent to suspension) did not alter
the total nuclear RNA synthesis, as shown by the total transcription ranging from 19-24 • 106 incorporated counts
per minute; (b) in the presence of 1 gg/ml of alpha-amanitin,
mRNA
LEVEL
Downloaded from on June 16, 2017
1.0
40
[ ."...'.
.,...'
"....'/
08
!!iiii
m
0.6
S/
1
0.4
0.:1
14
.//
:':/S
S/
....i..i;
//'d
collagen
collagen
II
x
.-.. 9
/
:
/
:
i/i/
10
IS: /
//
ii9
i.)!ii i .!!
," ." .9
/::/1
:" . / I
9' 9149.I
i
//'/
/ /
12
Figure 3. Levels of mRNAs
iii:?)i:i
/
,..'. . .
s
9 s.
//
./
:f/,
.....
"://
i.' /.. 9
L"9 . : 9
,10
..'..
9 ~
/
b" : / I
"L
!'/:1
0
2
4
WEEK
The Journal of Cell Biology, Volume 106, 1988
9
: .9
2
o
SUSPENSION
464
4
i
and gene transcription activity
for collagen types I, II, IX,
and X during chondrocyte differentiation in vitro. Vertical
bars represent the ratios between each value of suspension culture and the value
determined on the dedifferentiated cells, which was taken
as 1.0, for both mRNA level
and gene transcription activity. The relative amounts of
mRNAs were calculated from
densitometric analysis of different exposures of Figs. 1 and
2. Each filter was hybridized
with the specific probe pXCR7
and for each slot the value of
mRNA was normalized on the
basis of rRNA content. The
values of gene transcription
used are those reported in
Table I.
Published February 1, 1988
the incorporation of [32p]UTP into collagen RNA transcripts was inhibited by 75-80%, which indicated that these
transcripts resulted from RNA polymerase II activity (12).
A comparative summary of the evolution of mRNA levels
and transcription activity for the four collagen genes during
in vitro differentiation is given in Fig. 3. As can be seen, the
decrease in the rate of transcription of the gene encoding collagen type I was proportional to the decrease in mRNA level
at each time interval. Also for collagen types IX and X, the
extent of transcription activation was sufficient to account for
the accumulation of the respective mRNAs. In contrast, collagen type II mRNA accumulated in excess as compared with
the increase in transcriptional activity of the corresponding
gene.
lO0
80.
60_
40.
2o
Collagen H and X mRNA Levels at the Late Stage
of Differentiation of Chick Embryo Chondrocytes
0
'
'
2'
DAYS
Figure4. Levels of types II and X mRNAs in chondrocytes differentiated in vivo and completing their maturation in vitro. Total RNAs
were extracted from chondrocytes derived from 29-31 stage chick
embryo tibiae (0 d) and from the same cells cultured for 2, 7, and
21 d on agarose-coated dishes. The RNA was analyzed by Northern
blots (not shown) and the amount of RNA was estimated from densitometric scannings. Values are expressed in arbitrary units, assuming as 100 the highest value for each RNA. (Solidcircles) Type
X collagen mRNA; (open circles) type II collagen mRNA.
Discussion
In this paper we report studies on the steady-state levels of
collagen mRNAs and on the transcriptional activity of the
specific genes during the differentiation in vitro of chicken
hypertrophic chondrocytes starting from adherent dedifferentiated cells. The data presented elucidate some new aspects
of the expression of collagen genes during this process.
Type II collagen mRNA dramatically increases within the
first week after the transfer of the cells to suspension and
reaches its maximal level in the second week. During limb
development and in micromass cultures in vitro an increase
in the amount of type II collagen mRNA is concomitant with
the condensation phase of mesenchymal prechondrogenic
cells (10). The condensation phase is considered to be the
first major event during chondrogenesis. It is noteworthy that
in our culture system adherent dedifferentiated cells undergo
an aggregation phase immediately after their transfer to suspension.
The change in the transcriptional activity of type II collagen gene has the same pattern as the one of its messenger
RNA, reaching a peak at the second week; however, the
higher level of mRNA as compared with the rate of transcription suggests that an additional stabilization of this mRNA
could also take place during differentiation. Accumulation of
type IX collagen mRNA follows a pattern similar to the one
of the mRNA for collagen type II. We observed that the increase in mRNA for collagen type IX is mostly due to transcription activation, although the slight discrepancy between
rate of transcription and level of mRNA at the fourth week
could indicate an accelerated degradation of the mRNA.
Mueller-Glauser et al. (16) have recently shown that type
IX collagen is localized in the intersections of type II collagen fibrils and probably contributes to their stabilization. It
Castagnola et al.
Collagen Expression in Differentiating Chondrocytes
is tempting to speculate that the simultaneous increase of
types II and IX mRNAs reflects the necessity of a correct assembly of extracellular fibrils for the contemporary occurrence of both collagens.
Type X collagen mRNA increase is more continuous and
progressive and reaches its maximum at 3-4 wk, when essentially all cultured cells are hypertrophic. At a later stage
of cell differentiation a decrease is also observed in type X
collagen mRNA (see Fig. 4). This could be due both to an
additional specific regulation mechanism of collagen expression or to a senescence phenomena. The accumulation of
type X collagen mRNA by differentiating hypertrophic chondrocytes is specifically inhibited by the presence of 180 mM
dimethylsulfoxide in the culture medium (15). We can now
ask which stimulus induces in chondrocytes the transcription
of type X collagen mRNA both in vivo and in vitro. Microenvironment changes are certainly involved. The culture
conditions may play a relevant role since chondrocytes from
the permanent cartilaginous caudal zone of chicken sternum, when grown on agarose-coated dishes, also differentiate to hypertrophic chondrocytes synthesizing type X collagen (4, 21).
Our results further indicate that during the in vitro differentiation process the level of type I collagen mRNA (relatively high in the adherent dedifferentiated cells) rapidly
decreases, though a detectable amount is still present in well
differentiated chondrocytes. These results are in agreement
with those obtained in vivo, where both the switch from synthesis of type I to type II collagen and the presence of low
levels of type I collagen mRNA in already differentiated
chondrocytes were observed (26). A translational downregulation of type I collagen during the chondrogenesis pro-
465
Downloaded from on June 16, 2017
To further investigate the evolution of types II ~nd X collagen
mRNAs at a late stage of differentiation, chondrocytes
freshly dissociated from stage 30 chick embryo tibiae were
directly cultured in suspension. Total RNA was extracted
from the tissue itself as from cells cultured for 2, 7, and 21 d.
Fig. 4 shows the levels of these mRNAs, estimated from
scanning of the Northern blots. A continuous decrease of the
amount of the messenger for collagen type II was observed,
whereas the amount of collagen type X increased to reach
a peak at 7 d and decreased thereafter.
Published February 1, 1988
The Journal of Cell Biology, Volume 106, 1988
transition from a stage characterized by a high level of type
I collagen mRNA to a stage characterized by a predominance
of types II and IX collagen mRNAs and (b) a transition to
a stage characterized by the highest level of type X collagen
mRNA.
Beatrice Dozin is the recipient of a long-term European Molecular Biology
Organization fellowship. We thank Drs. E. Avvedimento, M. Sobel, Y.
Ninomiya, and E Amaldi for kindly providing cDNA probes.
This work was supported by grants from Progetti Finalizzati Oncologia
and Ingegneria Genetica e Basi Molecolari delle Malattie Ereditarie, Consiglio Nazionale delle Ricerche, Rome, and by funds from Ministero Pubblica Istruzione, Italy.
Received for publication 10 August 1987, and in revised form 29 September
1987.
References
I. Ambesi-lmpiombato, F. S., L. A. Parks, and H. G. Coon. 1980. Culture
of hormone-dependent functional epithelial cells from rat thyroids. Proc.
Natl. Acad. Sci. USA. 77:3455-3459.
2. Capasso, O., G. Tajana, and R. Cancedda. 1984. Location of 64K collagen
producer chondrocytes in developing chick embryo tibiae. Mol. Cell.
Biol. 4:1163-1168.
3. Castagnola, P., G. Moro, F. Descalzi-Cancedda, and R. Cancedda. 1986.
Type X collagen synthesis during in vitro development of chick emb~o
tibiae chondrecytes. J. Cell Biol. 102:2310-2317.
4. Castagnola, P., G. Torella, and R. Cancedda. 1987. Type X collagen synthesis by cultured chondrocytes derived from the permanent cartilaginous
region of chick embryo sternum. Dev. Biol. 123:332-337.
5. Clayton, D. F., and J. E. Darnell, Jr. 1983. Changes in liver-specific compared to common gene transcription during primary culture of mouse hepatocytes. Mol. Cell. Biol. 3:1552-1561.
6. Duch6ne, M., M. E. Sobel, and P. K. Mueller. 1982. Levels of collagen
mRNA in dedifferentiating chondrocytes. Exp. Cell Res. 142:317-324.
7. Erie, D. E. 1983. Cellular condensation and chondrogenesis. In Cartilage.
Vol. 2. B. K. Hall, editor. Academic Press, Inc., New York. 143-156.
8. Focht, R. V., and S. L. Adams. 1984. Tissue specificity of type I collagen
gene expression is determined at both transcriptional and posttranscriptional levels. Mol. Cell. Biol. 4:1843-1852.
9. Hamburger, V., and H. L. Hamilton. 1951. A series of normal stages in
the development of the chick embryo. J. Morphol. 88:49-92.
10. Kosher, R. A., W. M. Kulyk, and S. W. Gay. 1986. Collagen gene expression during limb cartilage differentiation. J. Cell Biol. 102:1151-1156.
1 I. Kravis, D., and W. B. Upholt. 1985. Quantitation of type II procollagen
mRNA levels during chick limb cartilage differentiation. Dev. Biol. 108:
164-172.
12. Lindell, T. J., F. Weinberg, P. W. Morris, R. G. Roeder, and W. J. Rutter.
1970. Specific inhibition of nuclear RNA polymerase I1 by alphaamanitin. Science (Wash. DC). 170:447--449.
13, Maniatis, T., E. F. Fritseh, and J. Sambrook. 1982. Molecular Cloning.
Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 196-204.
14. McKnight, G. S., and R. D. Palmiter. 1979. Transcriptional regulation of
the ovalbumin and conalbumin genes by steroid hormone in chick
oviduct. J. Biol. Chem. 254:9050-9058.
15. Manduca, P., P. Castagnola, and R. Cancedda. 1988. Dimethylsulfoxide
interferes with in vitro differentiation of chick embryo endochondral
chondrocytes. Dev. Biol. In press.
16. Mueller-Glauser, W., B. Humbel, M. Glatt, P. Straeuli, K. H. Winterhalter, and P. Bruckner. 1986. On the role of type IX collagen in the
extracellular matrix of cartilage: type IX collagen is localized to intersections of collagen fibrils. J. Cell Biol. 102:1931-1939.
17. Ninomiya, Y., M. van der Rest, R. Mayne, G. Lozano, and B. R. Olsen.
1985. Construction and characterization of cDNA encoding the alpha 2
chain of chicken type IX collagen. Biochemistry. 24:4223-4229.
18. Ornoy, A., and I. Zusman. 1983. Vitamins and cartilage. In Cartilage. Vol.
2. B. K. Hall, editor. Academic Press, Inc., New York. 298-300.
19. Saxe, S. A., L. N. Lukens, and P. J. Pawlowski. 1985. Changes in the nuclear and cytoplasmatic levels of type I and type II collagen RNAs during
growth of chondrocytes in 5-bromo-2'-deoxyuridine. J. Biol. Chem. 260:
3812-3819.
20. Schmid, T. M., and T. F. Linsenmayer. 1985. Immunohistochemicallocalization of short chain cartilage collagen (type X) in avian tissues. J. Cell
Biol. 100:598-605.
21. Solursh, M., K. L. Jensen, R. S. Reiter, T. M. Schmid, and T. F. Linsenmayer. 1986. Environmental regulation of type X collagen production by
cultures of limb mesenchyme, mesectoderm, and sternal chondrocytes.
Dev. Biol. 117:90-101.
22. Solursh, M., K. L. Jensen, C. T. Singley, T. F. Linsenmayer, and R. S.
Reiter. 1982. Two distinct regulatory steps in cartilage differentiation.
466
Downloaded from on June 16, 2017
cess has been reported by other investigators (8, 10, 19, 11).
In this paper we show that the transcriptional activity of the
type I collagen gene decreased with differentiation. Therefore, the regulation of this messenger RNA occurs at the
level of transcription. In vivo, low but detectable levels of
type II collagen mRNA are present in mesenchymal cells at
stage 20/22 of limb development, well before overt cartilage
differentiation (11), and it has been suggested that a low level
of expression of type II collagen gene represents a molecular
marker of the state of determination of progenitor cells (10).
In agreement with this finding, our results indicate low but
detectable transcription activity and mRNA levels for
cartilage-specific collagen in the dedifferentiated cells. It
cannot be ruled out, however, that a small number of contaminating differentiated chondrocytes are responsible for
the presence of detectable types II, IX, and especially X collagen mRNAs.
After 2 wk of culture the amount of types II and IX collagen mRNAs in chondrocytes decreases. A decrease of type
II collagen mRNA has been also reported by Kravis and Upholt (11) in prolonged culture of chondrocytes; nevertheless,
given their culture conditions, it is likely that the decrease
they observed was due, at least partially, to a dedifferentiation of chondrocytes.
It is important to mention here that we have several lines
of evidence suggesting that, in our culture system, the first
stage concerned with the high expression of collagen type I
as well as the subsequent reversion to the synthesis of
cartilage-specific collagens might reflect the in vivo situation. (a) The dedifferentiated cells were obtained by enzymatic digestion of whole tibiae of stage 28/30 chicken embryo. Such cell population used as a starting point of the in
vitro culture is hetergeneous in terms of stage of differentiation; nevertheless, all cells assume the fibroblast phenotype
and revert to type I collagen synthesis. (b) Cloned cells obtained by limiting dilution of the freshly dissociated 6-d embryo tibia, when grown in anchorage-dependent conditions
(fibroblast-like morphology) and transferred thereafter in
suspension culture in the constant presence of ascorbic acid,
are able to differentiate to form a structure analogous to
hypertrophic cartilage (23). (c) We have evidence that cultured prechondrogenic limb bud cells can undergo a sequence of morphological and biochemical events almost superimposable to those described with dedifferentiated cells
(Tachetti, C., B. Dozin, and R. Cancedda, manuscript in
preparation).
We have also to point out that in our culture system the decrease in the types II and IX collagen mRNAs is concomitant
with the achievement of the highest type X collagen mRNA
level.
Given all that, we suppose that during chondrocyte
differentiation, after the first shift in the collagens synthesized at the time chondrogenesis begins, a second shift occurs between type II (and IX) and type X collagen when
chondrocytes become hypertrophic.
In summary, based on the data presented in this paper and
on our previous observations on collagen synthesis (3) and
kinetic properties of chicken chondrocyte undergoing differentiation (Giaretti, W., G. Moro, R. Quarto, S. Bruno, A.
DiVinci, E. Geido, and R. Cancedda, manuscript submitted
for publication), we suggest that chicken chondrocyte differentiation proceeds through at least two different steps: (a) a
Published February 1, 1988
Dev. Biol. 94:311-325.
23. Tacchetti, C., R. Quarto, L. Nitsch, D. J. Hartmann, and R. Cancedda.
1987. In vitro morphogenesis of chick embryo hypertrophic cartilage. J.
Cell Biol. 105:999-1006.
24. Thomas, P. S. 1980. Hybridization of denatured RNA and small DNA fragments transferred to nitrocellulose. Proc. Natl. Acad. Sci. 77:52015205.
25. Tullis, R. H., and H. Rubin. 1980. Calcium protects DNase I from proteinase K: a new method for the removal of contaminating RNase from
DNase I. Anal. Biochem. 107:260-264.
26. von der Mark, K. 1986. Differentiation, modulation and dedifferentiation
of chondrocytes. Rheumatology. 10:272-315.
27. yon der Mark, K., and H. yon der Mark. 1977. Immunological and biochemical studies of collagen type transition during in vitro chondrogenesis of chick limb mesodermal cells. J. Cell Biol. 73:736-747.
28. Yamamoto, T., M. E. Sobel, S. L. Adams, V. E. Avvedimento, R. Di
Lauro, I. Pastan, B. de Crombrugghe, A. Showalter, D. Pesciotta, P.
Fietzek, and B. R. Oisen. 1980. Construction of a recombinant bacterial
plasmid containing Pro-al(1) collagen DNA sequences. J. Biol. Chem.
255:2612-2615.
29. Young, M. F., G. Vogeli, A. M. Nunez, M. P. Fernandez, M. Sullivan,
and M. Sobel. 1984. Isolation of eDNA and genomic DNA clones encoding type II collagen. Nucl. Acids Res. 12:4207-4228.
Downloaded from on June 16, 2017
Castagnola et al. Collagen Expression in Differentiating Chondrocytes
467